Pith. sign in

REVIEW 1 major objections 8 minor 60 references

Embracing Spillover: Spatial Effects in Experiments

T0 review · 1 major / 8 minor · reviewed 2026-07-08 · glm-5.2

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read Spillover level can't be learned from cluster trials — only its shape

desk verdict Clean framework for spatial interference in cluster trials; core math is sound but abstract overstates scope and proofs/code are missing. read the letter →

arxiv 2607.06506 v1 pith:L4CWZ4F7 submitted 2026-07-07 stat.ME

Samuel I. Watson , Penny Hancock , Jack Fraser-Govil , Kieran DeWalt , Thomas A. Smith This is my paper
classification stat.ME
keywords levelbiaseffectsexposurekernelanchoranchoringdata
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper proves that in cluster-randomized experiments with spatial spillover, the data can identify the shape of the spillover kernel (how effects vary with distance) but never its overall level (how much spillover there is in total). This non-identifiability is structural: it holds at every sample size, under every outcome model in the linear-predictor class, and cannot be broken by better statistical modeling of variance or random effects. Every estimator therefore silently relies on an anchoring assumption that fixes the level, and the paper derives an exact three-part bias decomposition for any such anchored estimator: a level term (proportional to how wrong the anchor is), a shape error (from misspecifying the kernel's form), and a leakage term (absorbed by the realized spatial geometry). A single scalar called the level leverage measures each causal estimand's exposure to the unidentified level, giving the exact level bias, the variance cost of estimating the level rather than assuming it, and the mean-squared-error floor that no additional data can remove. The conventional cluster-trial difference-in-means estimator is shown to be the special case of an implicit anchor whose contamination bias is given in closed form. Existing approaches — Bernoulli randomization, elicited decay bounds, assumed compact support — are all recast as different anchoring choices within this single framework, and the paper provides a taxonomy of design modifications that can purchase the missing level information, along with a criterion for when augmenting the design beats simply accepting an anchor.

What carries the argument

The level leverage ℓ(w,c) = m(w) + qc, a scalar that counts the per-unit exposure pairs through which an estimand references the unexposed counterfactual. It simultaneously gives the direction of non-identifiability, the exact first-order bias of any anchored estimator, the variance premium for estimating the level instead, and the MSE floor that persists as sample size grows.

What would settle it

If a cluster-randomized trial with fixed margins were augmented with a design feature generating residual variation in the constant-exposure vector (e.g., sentinel units with zero exposure), and the estimated level from that augmentation disagreed with the level implied by an anchor assumed valid under the canonical design, the anchor would be refuted.

Watch

Extended reading notes

Core claim

Under cluster randomization with fixed exposure set size, the level of the spillover kernel is aliased with the intercept and direct effect at every sample size and under every outcome model in the linear-predictor class. The data identify only the centered kernel (its shape), while every causal estimand whose level leverage is nonzero — including the direct effect, total cluster effect, and dose-response curve — depends on the unidentified level and therefore requires an anchoring assumption. The bias of any anchored plug-in decomposes exactly into a level term, a shape error, and a leakage term, each computable from the design before data collection. The conventional cluster-trial analysis

Load-bearing premise

The entire framework assumes the spillover effect on each individual is a sum of contributions from nearby treated sources, each weighted by a function of distance — the so-called additive linear exposure mapping. If the true exposure-response relationship is nonlinear in the count of treated sources (for example, if two nearby sources interact synergistically rather than additively), the level-aliasing structure and the closed-form bias decomposition may not hold as stated.

Editorial extensions

If this is right

  • Trialists can compute, before unblinding, exactly which estimands are vulnerable to the unidentified level and by how much, enabling informed design choices rather than post-hoc damage control.
  • The traditional practice of spacing clusters apart to prevent spillover is precisely the design that makes the no-spillover assumption unfalsifiable — it removes the variation needed to test it.
  • Design augmentations such as sentinel units, baseline periods, or randomised saturation can be compared on a common scale via the level information I₀, giving a principled criterion for when to augment versus anchor.
  • Estimated effects from spaced cluster trials may not transport to new geographic configurations, because the realized geometry is baked into the implicit anchor.
  • Distance contrasts of the spillover curve (effects at one distance minus effects at another) are identified by randomization alone and require no anchoring assumption, making them the robust estimands of choice when the level cannot be credibly anchored.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit.

Referee Report

1 major / 8 minor

Summary. This manuscript develops a framework for spatial experiments with spillover effects under cluster randomization. The central result (Proposition 1) is that under a fixed exposure set size (H1-H3), the level of the spillover kernel is aliased with the intercept and direct effect at every sample size and within the linear-predictor model class. The paper derives an exact bias decomposition (Theorem 3) for any anchored plug-in estimator into a level term, a shape error, and a leakage term, and introduces the level leverage scalar measuring each estimand's exposure to the unidentified level. Corollary 1 specializes the result to the conventional cluster-trial analysis. The paper provides a taxonomy of design augmentations that can identify the level and a decision criterion for when to augment versus anchor. A simulation study confirms the asymptotic predictions.

Significance. The paper makes a substantive conceptual contribution by unifying several existing approaches (Bernoulli randomization, compact support assumptions, buffer designs) as special cases of an anchoring choice within a single framework. The level leverage scalar is a genuinely useful design-computable diagnostic, and the exact variance decomposition in Proposition 4(i) and the MSE floor result (Corollary 2) are clean, falsifiable predictions. The taxonomy in Table 4 is practically valuable for trialists. The mathematical arguments are sound: Proposition 1's level aliasing is a direct algebraic observation, Theorem 3 follows from GLS projection structure, and Corollary 1's contamination bias formula is a clean specialization. The paper is appropriately honest about the scope limitations of the linear exposure mapping in Section 7.

major comments (1)
  1. Abstract and Proposition 1(iv) scope mismatch. The abstract states the level is aliased 'at any sample size and under any outcome model.' Proposition 1(iv) actually covers the class where the law of y given z depends on (alpha, tau, phi) only through eta(z) = alpha*1 + tau*z_tilde + S[phi](z), which includes linear models with arbitrary Sigma, GLMs, and mixed models with conditional law depending on eta + Zb. This does not cover models where the exposure enters nonlinearly in the treated-source count (e.g., saturating dose-response or threshold effects), which are scientifically plausible for the motivating examples of vector-borne disease transmission and gene drives. The abstract phrasing should be corrected to accurately reflect the linear-predictor class scope of Proposition 1(iv). This is a load-bearing issue because the central claim's reach is being overstated in the paper's most-
minor comments (8)
  1. Section 1.1, paragraph 2: 'inclduing' should be 'including' in the abstract.
  2. Section 2.1: The notation z_{k(i)} and z_{iota(t)} are both used for cluster treatment status; a brief clarifying remark that these refer to the same quantity under different indexing would help readers.
  3. Table 2: The column header 'c_w' appears to be 'c' (the direct-effect indicator) and 'w' (the measure); consider labeling more explicitly as '(c, w)' or adding a footnote.
  4. Section 4.3: The definition of the parametric working class K references 'b psi_p(.; rho)' but the notation 'b' as both a coefficient and a function prefix is slightly confusing; consider using a different symbol for the coefficient.
  5. Section 5.2: 'preiod' should be 'period' (appears twice in the paragraph containing 'the intervention preiod of the study').
  6. Figure 1: The y-axis labels and legend are small; consider enlarging for readability in print format.
  7. Section 6: The simulation study would benefit from stating the number of Monte Carlo replications used.
  8. Section 4.4, Condition 1: The role of the norm ||.||_R is introduced but its specific choice is not discussed; a brief remark on practical selection would be helpful.

Simulated Author's Rebuttal

1 responses · 0 unresolved

The referee identifies a scope mismatch between the abstract's claim that level aliasing holds 'under any outcome model' and the actual scope of Proposition 1(iv), which covers the linear-predictor model class (linear models with arbitrary covariance, GLMs, and mixed models whose conditional law depends on parameters through the linear predictor). The referee correctly notes that models with nonlinear exposure in the treated-source count—such as saturating dose-response or threshold effects—are not covered and are scientifically plausible for the motivating examples. We agree this is a load-bearing issue: the abstract overstates the reach of the central claim and must be corrected.

read point-by-point responses

  1. Referee: Abstract and Proposition 1(iv) scope mismatch. The abstract states the level is aliased 'at any sample size and under any outcome model.' Proposition 1(iv) actually covers the class where the law of y given z depends on (alpha, tau, phi) only through eta(z) = alpha*1 + tau*z_tilde + S[phi](z), which includes linear models with arbitrary Sigma, GLMs, and mixed models with conditional law depending on eta + Zb. This does not cover models where the exposure enters nonlinearly in the treated-source count (e.g., saturating dose-response or threshold effects), which are scientifically plausible for the motivating examples of vector-borne disease transmission and gene drives. The abstract phrasing should be corrected to accurately reflect the linear-predictor class scope of Proposition 1(iv). This is a load-bearing issue because the central claim's reach is being overstated in the paper's most-

    Authors: The referee is correct on all counts. The abstract phrase 'under any outcome model' overstates the scope of Proposition 1(iv), which covers the linear-predictor class: models where the conditional law of y given z depends on (alpha, tau, phi) only through eta(z) = alpha*1 + tau*z_tilde + S[phi](z). This includes linear models with arbitrary Sigma, GLMs with E(y_i|z) = g^{-1}{eta_i(z)}, and mixed models whose conditional law given random effects depends on parameters through eta(z) + Zb. It does not cover models where the exposure enters nonlinearly in the treated-source count—for example, saturating dose-response of the form f(S[phi](z)) for nonlinear f, or threshold effects. These are scientifically plausible: vector-borne disease transmission often exhibits saturating or threshold dynamics in exposure to infectious agents, and gene drive release dynamics are inherently nonlinear in the treated-source count. We will correct the abstract to read 'at any sample size and within the linear-predictor model class' (or equivalent precise language), replacing 'under any outcome model.' We will also add a sentence to the abstract noting that the linear-predictor class encompasses linear models, GLMs, and mixed models, and that nonlinear exposure mappings are discussed as a scope limitation in Section 7. The body text is already accurate: Proposition 1(iv) states its scope precisely, and Section 7 explicitly acknowledges that 'our exact results are computed within the linear exposure mapping' and that outside it 'the closed-form prices... do not survive as stated.' The mismatch is confined to the abstract's shorthand, which we will fix. revision: yes

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity; central results derive from linear algebra of the exposure mapping and GLS projection properties, with one minor self-citation for a special case.

full rationale

The paper's central results are derived from first principles within the manuscript. Proposition 1 (level aliasing) follows from the linear algebra of the exposure mapping under H1-H3: S[1](z) = M(z)1 - q z̃ lies in col{[1, z̃]} by direct construction, so the flat direction (α-aM₀, τ+aq, ϕ+a) is a definitional consequence, not a fitted input renamed as output. Theorem 3 (bias decomposition) follows from GLS projection properties: the projection functional P_{N,w,c}(e;z) decomposes into level, shape, and leakage terms by standard linear algebra. The level leverage ℓ(w,c) = m(w) + qc is a design-computable scalar from the estimand definition and design constants, not a parameter fitted to data and then 'predicted.' Corollary 1 (CRT contamination bias) is derived as the special case of Theorem 3 with empty working class and between-cluster mean anchor, yielding E[τ̂_crt] = τ + (m₀-q)ϕ̄_w - m₀ϕ̄_b by substitution. The paper cites Watson and Smith (2025) for the compact-support anchor, but treats it as one anchoring choice within the framework (Table 4, §7), not as a load-bearing premise for the main theorems. The simulation study (§6) confirms theoretical predictions against external data-generating processes (Matérn and Wendland kernels), not against fitted values. The paper is transparent about scope: §7 explicitly states the closed-form prices 'do not survive as stated' outside the linear model. No step in the derivation chain reduces to its own inputs by construction; the self-citation is minor and non-load-bearing for the central claims. Score 1 reflects the minor self-citation without independent verification, which does not undermine the self-contained derivation.

Assumptions & free parameters 5 free parameters · 7 assumptions · 3 invented entities

The framework introduces three derived quantities (level leverage, level information, exposure perturbation) that are all algebraic consequences of the model structure and design, not postulated entities. The free parameters (alpha, tau, phi, rho, Sigma) are standard in spatial mixed models. The key axioms are the linear exposure mapping and conditions H1-H3, which are clearly stated and their consequences are honestly traced. No new physical entities or forces are postulated.

free parameters (5)
  • alpha (intercept)
    Model intercept in Eq. 1; estimated from data but aliased with kernel level under H1-H3
  • tau (direct effect)
    Direct treatment effect in Eq. 1; estimated from data but aliased with kernel level under H1-H3
  • phi (spillover kernel)
    Nonparametric distance kernel; shape is identified but level is not under H1-H3
  • rho (kernel lengthscale)
    Pseudo-true lengthscale for parametric working kernel families; profiled in estimation
  • Sigma parameters (spatial covariance)
    Stationary isotropic spatial covariance parameters; estimated under C6
assumptions (7)
  • domain assumption Additive linear exposure mapping: Y_i(z) = alpha + tau*z_{k(i)} + S[phi]_i(z) + epsilon (Eq. 1)
    Central modeling assumption in Section 2.4; all identification and bias results are derived within this class. Nonlinear exposure mappings are excluded.
  • domain assumption SUTVA generalization: potential outcomes Y_i(x, a_i) depend on own treatment x and source configuration a_i, not on other units' outcomes directly
    Section 2.3; standard causal inference assumption extended to allow interference through exposure mapping
  • domain assumption H1 (Global exposure): every source lies within modelled range of every unit
    Section 2.2; structural condition for level aliasing in Proposition 1(ii)
  • domain assumption H2 (Homogeneous exclusion): A_i contains only own-cluster sources with |A_i|=q for all i
    Section 2.2; structural condition for level aliasing in Proposition 1(ii)
  • domain assumption H3 (Fixed margin): M(z) = M_0 for all z in Z
    Section 2.2; ensures flat direction is common to all allocations in Proposition 1(ii)
  • domain assumption Stationary isotropic spatial covariance (Sigma)
    Section 4.1; assumed for the working model. Relaxation to anisotropic or non-stationary covariance is not discussed.
  • standard math Conditions C2-C7 (regularity conditions)
    Section 4.4; standard regularity conditions for geometry stabilization, well-conditioned precision, covariance estimability, and CLT. Stated as provided in Supplementary Material.
invented entities (3)
  • Level leverage l(w,c) = m(w) + q*c independent evidence
    purpose: Scalar measuring each estimand's exposure to the unidentified kernel level
    Derived from the model structure (Proposition 1(iii)); computable from design before data collection; falsifiable in the sense that different designs yield different values
  • Level information I_0(z) independent evidence
    purpose: Measures Omega-weighted variation in constant-exposure vector after working columns partialled out
    Derived from design matrices (Proposition 4); computable from candidate geometry before unblinding; determines whether level can be estimated vs anchored
  • Exposure perturbation r(z) = s_0(z) - M_0*1 + q*z_tilde independent evidence
    purpose: Characterizes design features that generate level-identifying variation
    Algebraic decomposition of the constant-exposure vector; each design augmentation in Table 4 generates a specific form of r(z)

how reviews work

0 comments
Cite this review

Pith. "Pith review of Embracing Spillover: Spatial Effects in Experiments." pith.science (2026). https://pith.science/paper/L4CWZ4F7

@misc{pith2026260706506,
  author       = {Pith},
  title        = {Pith review of: Embracing Spillover: Spatial Effects in Experiments},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L4CWZ4F7}},
  note         = {Machine review of arXiv:2607.06506}
}
read the original abstract

Interventions delivered in space generate effects that spill over between experimental units. We develop a framework for spatial experiments in which causal estimands, inclduing direct, indirect, total, and dose--response effects, are linear functionals of a spillover kernel. Under cluster randomisation with a fixed exposure set size, the data identify the shape of the kernel but not its level: the level is aliased with the intercept and direct effect, at any sample size and under any outcome model. Every estimator therefore rests on an anchoring assumption that fixes the level. We derive an exact decomposition of the bias of any anchored estimator into a level term, a shape error from kernel misspecification, and a leakage term absorbed by the realised geometry, each computable from the design before data collection. A single scalar, the level leverage, gives each estimand's exposure to the unidentified level, the exact level bias, and the variance cost of estimating the level instead. The conventional cluster-trial analysis is the special case of an implicit anchor, with contamination bias in closed form. Existing approaches, including identification through Bernoulli randomisation, elicited bounds on interference decay, and assumed compact support, are, within the linear exposure mapping, anchoring choices in this framework. We give a taxonomy of design augmentations that purchase the level and a criterion for when to augment and when to anchor.

Figures

Figures reproduced from arXiv: 2607.06506 by the authors.

Figure 1
Figure 1. Example design stage diagnostics for a cluster randomised trial design on |e¯| settles the anchor-or-identify decision for every level-dependent estimand in [PITH_FULL_IMAGE:figures/full_fig_p020_1.png] view at source ↗
Figure 2
Figure 2. Simulation study results showing bias (top panel), coverage (middle panel) and mean squared error (bottom) panel [PITH_FULL_IMAGE:figures/full_fig_p022_2.png] view at source ↗

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

60 extracted references · 60 canonical work pages

  1. [1]

    Vanderweele and Guanglei Hong and Stephanie M

    Tyler J. Vanderweele and Guanglei Hong and Stephanie M. Jones and Joshua L. Brown , doi =. Mediation and Spillover Effects in Group-Randomized Trials: A Case Study of the 4Rs Educational Intervention , volume =. Journal of the American Statistical Association , month =

    work page

  2. [2]

    Hilbert space methods for reduced-rank Gaussian process regression , volume =

    Arno Solin and Simo Särkkä , doi =. Hilbert space methods for reduced-rank Gaussian process regression , volume =. Statistics and Computing , month =

    work page

  3. [3]

    Aronow and Michael G

    Fredrik Sävje and Peter M. Aronow and Michael G. Hudgens , doi =. Average treatment effects in the presence of unknown interference , volume =. The Annals of Statistics , month =

    work page

  4. [4]

    Aronow and Cyrus Samii , doi =

    Peter M. Aronow and Cyrus Samii , doi =. Estimating average causal effects under general interference, with application to a social network experiment , volume =. The Annals of Applied Statistics , month =

    work page

  5. [5]

    Jacklin F Mosha and Nancy S Matowo and Manisha A Kulkarni and Louisa A Messenger and Eliud Lukole and Elizabeth Mallya and Tatu Aziz and Robert Kaaya and Boniface A Shirima and Gladness Isaya and Monica Taljaard and Ramadhan Hashim and Jacklin Martin and Alphaxard Manjurano and Immo Kleinschmidt and Franklin W Mosha and Mark Rowland and Natacha Protopopof...

    work page

  6. [6]

    Hussey and James P

    Michael A. Hussey and James P. Hughes , doi =. Design and analysis of stepped wedge cluster randomized trials , volume =. Contemporary Clinical Trials , month =

    work page

  7. [7]

    Glass and Fiona Vanobberghen and Thomas Smith , doi =

    Lea Multerer and Tracy R. Glass and Fiona Vanobberghen and Thomas Smith , doi =. Analysis of contamination in cluster randomized trials of malaria interventions , volume =. Trials , month =

    work page

  8. [8]

    Watson and Yixin Wang and Emanuele Giorgi , month =

    Samuel I. Watson and Yixin Wang and Emanuele Giorgi , month =. Approximate Likelihood-Based Inference for Spatial Generalized Linear Mixed Models , year =

    work page

Show all 60 references
  1. [9]

    Spatiotemporal causal inference with arbitrary spillover and carryover effects: Airstrikes and insurgent violence in the Iraq War , year =

    Mitsuru Mukaigawara and Kosuke Imai and Jason Lyall and Georgia Papadogeorgou , month =. Spatiotemporal causal inference with arbitrary spillover and carryover effects: Airstrikes and insurgent violence in the Iraq War , year =

  2. [10]

    Causal inference with misspecified exposure mappings: separating definitions and assumptions , volume =

    F Sävje , doi =. Causal inference with misspecified exposure mappings: separating definitions and assumptions , volume =. Biometrika , month =

  3. [11]

    Design-based inference for spatial experiments under unknown interference , volume =

    Ye Wang and Cyrus Samii and Haoge Chang and PM Aronow , doi =. Design-based inference for spatial experiments under unknown interference , volume =. The Annals of Applied Statistics , month =

  4. [12]

    Estimands in cluster-randomized trials: choosing analyses that answer the right question , volume =

    Brennan C Kahan and Fan Li and Andrew J Copas and Michael O Harhay , doi =. Estimands in cluster-randomized trials: choosing analyses that answer the right question , volume =. International Journal of Epidemiology , month =

  5. [13]

    Compactly Supported Correlation Functions , volume =

    Tilmann Gneiting , doi =. Compactly Supported Correlation Functions , volume =. Journal of Multivariate Analysis , month =

  6. [14]

    Kilpatrick and Michael G

    Kayla W. Kilpatrick and Michael G. Hudgens and M. Elizabeth Halloran , doi =. Estimands and inference in cluster‐randomized vaccine trials , volume =. Pharmaceutical Statistics , month =

  7. [15]

    Twenty‐five years of confirmatory adaptive designs: opportunities and pitfalls , volume =

    Peter Bauer and Frank Bretz and Vladimir Dragalin and Franz König and Gernot Wassmer , doi =. Twenty‐five years of confirmatory adaptive designs: opportunities and pitfalls , volume =. Statistics in Medicine , month =

  8. [16]

    Jarvis and Lea Multerer and Daniel Lewis and Fred Binka and W

    Christopher I. Jarvis and Lea Multerer and Daniel Lewis and Fred Binka and W. John Edmunds and Neal Alexander and Thomas A. Smith , doi =. Spatial Effects of Permethrin-Impregnated Bed Nets on Child Mortality: 26 Years on, a Spatial Reanalysis of a Cluster Randomized Trial , v...

  9. [17]

    Zimmerman , doi =

    Hao Zhang and Dale L. Zimmerman , doi =. Towards reconciling two asymptotic frameworks in spatial statistics , volume =. Biometrika , month =

  10. [18]

    Smith and Luzia N

    Thomas A. Smith and Luzia N. Felber and Neal A. Alexander and Pélagie E. B. Aboa and Claver N. Adjobi and Julien Z. B. Zahouli and Pie Müller , doi =. A diffusion model for mosquito control trials with spillover: application to calculations of power, sample size, and bias , ye...

  11. [19]

    Impact of spatial distribution of permethrin-impregnated bed nets on child mortality in rural northern Ghana

    F N Binka and F Indome and T Smith , doi =. Impact of spatial distribution of permethrin-impregnated bed nets on child mortality in rural northern Ghana. , volume =. The American Journal of Tropical Medicine and Hygiene , month =

  12. [20]

    Chennuri and Julia A

    Tibebu Habtewold and Dickson Wilson Lwetoijera and Astrid Hoermann and Rajabu Mashauri and Fatuma Matwewe and Rehema Mwanga and Prisca Kweyamba and Gilbert Maganga and Beatrice Philip Magani and Rachel Mtama and Moze Ally Mahonje and Mgeni Mohamed Tambwe and Felista Tarimo and...

  13. [21]

    Cluster Randomised Trials , year =

    RJ Hayes and LH Moulton , city =. Cluster Randomised Trials , year =

  14. [22]

    Airoldi and Fabrizia Mealli , doi =

    Laura Forastiere and Edoardo M. Airoldi and Fabrizia Mealli , doi =. Identification and Estimation of Treatment and Interference Effects in Observational Studies on Networks , volume =. Journal of the American Statistical Association , month =

  15. [23]

    Elizabeth Halloran , doi =

    Michael G Hudgens and M. Elizabeth Halloran , doi =. Toward Causal Inference With Interference , volume =. Journal of the American Statistical Association , month =

  16. [24]

    Leung , doi =

    Michael P. Leung , doi =. Causal Inference Under Approximate Neighborhood Interference , volume =. Econometrica , pages =

  17. [25]

    Watson , doi =

    Samuel I. Watson , doi =. Efficient design of geographically-defined clusters with spatial autocorrelation , url =. Journal of Applied Statistics , month =

  18. [26]

    Spatial regression and spillover effects in cluster randomized trials with count outcomes , volume =

    Karim Anaya‐Izquierdo and Neal Alexander , doi =. Spatial regression and spillover effects in cluster randomized trials with count outcomes , volume =. Biometrics , month =

  19. [27]

    Watson and Thomas A

    Samuel I. Watson and Thomas A. Smith , doi =. Design and Analysis of Randomized Trials to Estimate Spatio-Temporally Heterogeneous Treatment Effects , year =. Journal of the American Statistical Association , month =

  20. [28]

    John Edmunds , doi =

    Christopher Jarvis and Gian Luca Di Tanna and Daniel Lewis and Neal Alexander and W. John Edmunds , doi =. Spatial analysis of cluster randomised trials: a systematic review of analysis methods , volume =. Emerging Themes in Epidemiology , month =

  21. [29]

    Optimal adaptive promising zone designs , volume =

    Cyrus Mehta and Apurva Bhingare and Lingyun Liu and Pralay Senchaudhuri , doi =. Optimal adaptive promising zone designs , volume =. Statistics in Medicine , month =

  22. [30]

    Point estimation, confidence intervals, and P ‐values for optimal adaptive two‐stage designs with normal endpoints , volume =

    Jan Meis and Maximilian Pilz and Björn Bokelmann and Carolin Herrmann and Geraldine Rauch and Meinhard Kieser , doi =. Point estimation, confidence intervals, and P ‐values for optimal adaptive two‐stage designs with normal endpoints , volume =. Statistics in Medicine , month =

  23. [31]

    The Adaptive designs CONSORT Extension (ACE) statement: a checklist with explanation and elaboration guideline for reporting randomised trials that use an adaptive design , year =

    Munyaradzi Dimairo and Philip Pallmann and James Wason and Susan Todd and Thomas Jaki and Steven A Julious and Adrian P Mander and Christopher J Weir and Franz Koenig and Marc K Walton and Jon P Nicholl and Elizabeth Coates and Katie Biggs and Toshimitsu Hamasaki and Michael A...

  24. [32]

    promising zone

    Julia M. Edwards and Stephen J. Walters and Cornelia Kunz and Steven A. Julious , doi =. A systematic review of the “promising zone” design , volume =. Trials , month =

  25. [33]

    Turnbull , doi =

    Christopher Jennison and Bruce W. Turnbull , doi =. Adaptive sample size modification in clinical trials: start small then ask for more? , volume =. Statistics in Medicine , month =

  26. [34]

    Turnbull , doi =

    Christopher Jennison and Bruce W. Turnbull , doi =. Adaptive and nonadaptive group sequential tests , volume =. Biometrika , month =

  27. [35]

    A Practical Guide to Cluster Randomised Trials in Health Services Research , url =

    Sandra Eldridge and Sally Kerry , city =. A Practical Guide to Cluster Randomised Trials in Health Services Research , url =. doi:10.1002/9781119966241 , isbn =

    doi

  28. [36]

    Adaptive Designs for Clinical Trials of Drugs and Biologics Guidance for Industry , year =

    US Food and Drug Administration , journal =. Adaptive Designs for Clinical Trials of Drugs and Biologics Guidance for Industry , year =

  29. [37]

    Group Sequential Methods with Applications to Clinical Trials , year =

    Christopher Jennison and BW Turnbull , city =. Group Sequential Methods with Applications to Clinical Trials , year =

  30. [38]

    Demets and K

    David L. Demets and K. K. Gordon Lan , doi =. Interim analysis: The alpha spending function approach , volume =. Statistics in Medicine , month =

  31. [39]

    Mehta and Stuart J

    Cyrus R. Mehta and Stuart J. Pocock , doi =. Adaptive increase in sample size when interim results are promising: A practical guide with examples , volume =. Statistics in Medicine , month =

  32. [40]

    Bauer and K

    P. Bauer and K. Kohne , doi =. Evaluation of Experiments with Adaptive Interim Analyses , volume =. Biometrics , month =

  33. [41]

    Adaptive Group Sequential Designs for Clinical Trials: Combining the Advantages of Adaptive and of Classical Group Sequential Approaches , volume =

    Hans-Helge Müller and Helmut Schäfer , doi =. Adaptive Group Sequential Designs for Clinical Trials: Combining the Advantages of Adaptive and of Classical Group Sequential Approaches , volume =. Biometrics , month =

  34. [42]

    Proschan and Sally A

    Michael A. Proschan and Sally A. Hunsberger , doi =. Designed Extension of Studies Based on Conditional Power , volume =. Biometrics , month =

  35. [43]

    A tutorial on sample size calculation for multiple-period cluster randomized parallel, cross-over and stepped-wedge trials using the Shiny CRT Calculator , volume =

    Karla Hemming and Jessica Kasza and Richard Hooper and Andrew Forbes and Monica Taljaard , doi =. A tutorial on sample size calculation for multiple-period cluster randomized parallel, cross-over and stepped-wedge trials using the Shiny CRT Calculator , volume =. International...

  36. [44]

    Estimating intra-cluster correlation coefficients for planning longitudinal cluster randomized trials: a tutorial , volume =

    Yongdong Ouyang and Karla Hemming and Fan Li and Monica Taljaard , doi =. Estimating intra-cluster correlation coefficients for planning longitudinal cluster randomized trials: a tutorial , volume =. International Journal of Epidemiology , month =

  37. [45]

    Hemming and T

    K. Hemming and T. P. Haines and P. J. Chilton and A. J. Girling and R. J. Lilford , doi =. The stepped wedge cluster randomised trial: rationale, design, analysis, and reporting , volume =. BMJ , month =

  38. [46]

    Macharia and Jack Woodmansey and Robert W

    Emanuele Giorgi and Peter M. Macharia and Jack Woodmansey and Robert W. Snow and Barry Rowlingson , doi =. Maplaria: a user friendly web-application for spatio-temporal malaria prevalence mapping , volume =. Malaria Journal , month =

  39. [47]

    Hancock and Ace North and Adrian W

    Penelope A. Hancock and Ace North and Adrian W. Leach and Peter Winskill and Azra C. Ghani and H. Charles J. Godfray and Austin Burt and John D. Mumford , doi =. The potential of gene drives in malaria vector species to control malaria in African environments , volume =. Natur...

  40. [48]

    Leung , doi =

    Michael P. Leung , doi =. Cluster-Randomized Trials with Cross-Cluster Interference , year =. Journal of the American Statistical Association , month =

  41. [49]

    Ahmad and Warsito Tantowijoyo and Eggi Arguni and M

    Adi Utarini and Citra Indriani and Riris A. Ahmad and Warsito Tantowijoyo and Eggi Arguni and M. Ridwan Ansari and Endah Supriyati and D. Satria Wardana and Yeti Meitika and Inggrid Ernesia and Indah Nurhayati and Equatori Prabowo and Bekti Andari and Benjamin R. Green and Lau...

  42. [50]

    Reducing contamination risk in cluster-randomized infectious disease-intervention trials , volume =

    Robert S McCann and Henk van den Berg and Willem Takken and Amanda G Chetwynd and Emanuele Giorgi and Dianne J Terlouw and Peter J Diggle , doi =. Reducing contamination risk in cluster-randomized infectious disease-intervention trials , volume =. International Journal of Epid...

  43. [51]

    Watson and Alan Girling and Karla Hemming , doi =

    Samuel I. Watson and Alan Girling and Karla Hemming , doi =. Optimal study designs for cluster randomised trials: An overview of methods and results , volume =. Statistical Methods in Medical Research , month =

  44. [52]

    Evaluation of combinatorial optimisation algorithms for c-optimal experimental designs with correlated observations , volume =

    Samuel I Watson and Yi Pan , doi =. Evaluation of combinatorial optimisation algorithms for c-optimal experimental designs with correlated observations , volume =. Statistics and Computing , month =

  45. [53]

    Turnbull , doi =

    Christopher Jennison and Bruce W. Turnbull , doi =. Mid‐course sample size modification in clinical trials based on the observed treatment effect , volume =. Statistics in Medicine , month =

  46. [54]

    Thompson and Clemence Leyrat and Katherine L

    Jennifer A. Thompson and Clemence Leyrat and Katherine L. Fielding and Richard J. Hayes , doi =. Cluster randomised trials with a binary outcome and a small number of clusters: comparison of individual and cluster level analysis method , volume =. BMC Medical Research Methodol...

  47. [55]

    Justus Hofmeyr and Neil Moran and Sue Fawcus and Lumaan Sheikh and George Gwako and Alfred Osoti and Ashraf Aswat and Kristie-Marie Mammoliti and Kulandaipalayam N

    Ioannis Gallos and Adam Devall and James Martin and Lee Middleton and Leanne Beeson and Hadiza Galadanci and Fadhlun Alwy Al-beity and Zahida Qureshi and G. Justus Hofmeyr and Neil Moran and Sue Fawcus and Lumaan Sheikh and George Gwako and Alfred Osoti and Ashraf Aswat and Kr...

  48. [56]

    N. E. Breslow and D. G. Clayton , doi =. Approximate Inference in Generalized Linear Mixed Models , volume =. Journal of the American Statistical Association , month =

  49. [57]

    Completeness of reporting and risks of overstating impact in cluster randomised trials: a systematic review , volume =

    Elizabeth L Turner and Alyssa C Platt and John A Gallis and Kaitlin Tetreault and Christina Easter and Joanne E McKenzie and Stephen Nash and Andrew B Forbes and Karla Hemming and Christine Adrion and Naseerah Akooji and Hannah Bensoussane and Aneel Bhangu and Jon Bishop and B...

  50. [58]

    Choko and Elizabeth L

    Augustine T. Choko and Elizabeth L. Corbett and Nigel Stallard and Hendramoorthy Maheswaran and Aurelia Lepine and Cheryl C. Johnson and Doreen Sakala and Thokozani Kalua and Moses Kumwenda and Richard Hayes and Katherine Fielding , doi =. HIV self-testing alone or with additi...

  51. [59]

    Hemming and J

    K. Hemming and J. Thompson and C. Kristunas and A. Forbes and S. Watson and R. Bowden and J.E. McKenzie and J. Kasza , doi =. The performance of small sample correction methods for controlling type I error when analyzing parallel cluster randomized trials: a systematic review ...

  52. [60]

    Optimal allocation of observations in stepped-wedge and other cluster studies with correlated cluster-period effects , volume =

    A Girling and SI Watson , journal =. Optimal allocation of observations in stepped-wedge and other cluster studies with correlated cluster-period effects , volume =

Pith tools

Reviewed July 8, 2026 · model on record in the stance chip above.